Spurious Correlation
A failure mode in machine learning in which a model captures features that are statistically associated with labels but not causally related to them, and makes predictions based on those features. It degrades out-of-distribution generalization.
Timeline (last 90 days)
Spurious correlation is a common failure mode in machine learning: the model captures features that are statistically correlated with labels but not causally related, and makes predictions based on them.
In animal individual identification, the same animal frequently appears in front of fixed water sources, dens, or specific vegetation backgrounds, and the model may treat background texture or lighting conditions as identity cues.
AI research tools at the hypothesis-generation stage face two typical risks: spurious correlations and hallucinations.
All Facts (3)
Spurious correlation is a common failure mode in machine learning: the model captures features that are statistically correlated with labels but not causally related, and makes predictions based on them.
50%UnverifiedIn animal individual identification, the same animal frequently appears in front of fixed water sources, dens, or specific vegetation backgrounds, and the model may treat background texture or lighting conditions as identity cues.
50%UnverifiedAI research tools at the hypothesis-generation stage face two typical risks: spurious correlations and hallucinations.
50%